SPACE PERCEPTION OF TORTOISES.

By ROBERT M. YERKES.

(From the Harvard Psychological Laboratory, HUGO MÜNSTERBERG, Director.)

The Sense of Support in Animals.

A number of investigators have noticed that the young of many animals possess a sense of support, and that their behavior is adapted to the spatial conditions in which they happen to be placed. It is this sense of support that saves the sightless kitten or puppy from falls; but in case of the young chick which similarly hesitates when it approaches the edge of a void visual stimuli apparently determine the reaction. These reactions to spatial conditions are controlled by a complex of sense impressions which is still unanalyzed. In certain animals visual impressions seem to be all-important; in others organic data are chiefly significant, and again in other organisms there are indications of degrees of sensitiveness, if not modes of sense, of which we have no direct knowledge. And so, strange as it may seem, the "spatial worth" of sense data, as JAMES would call it, is no more a matter of accurate knowledge than is the development of the sense of space, or the modes of behavior in different spatial conditions exhibited by any animal.

THORNDIKE ('99, p. 284), who has studied the behavior of young chicks with reference to spatial relations, says "If one puts a chick on top of a box in sight of his fellows below, the chick will regulate his conduct by the height of the box." A chick 95 hours old does not hesitate to jump off at heights of 1 to 10 inches; at 22 inches it often hesitates a long time, and at 39 inches it usually does not jump at all. Furthermore,

immediately after hatching, young chicks are able to peck at objects with considerable accuracy, and they apparently estimate distances fairly well before they have had much experience outside the shell.

The behavior of young pigeons, chicks, kittens and puppies in unusual spatial conditions has been studied most fully by MILLS ('98, p. 150), who in discussing the "sense of support" writes: "I have found in the case of all puppies, and several other kinds of animals examined, that even on the first day of birth they will not creep off a surface on which they rest, if elevated some little distance above the ground. When they approach the edge they manifest hesitation, grasp with their claws or otherwise attempt to prevent themselves falling, and, it may be, cry out, giving evidence of some profound disturb. ance in their nervous system.

"It would seem that there is no more urgent psychic necessity to young mammals than this sense of being supported. All their ancestral experiences have been associated with terra firma, so that it is not very surprising that when terra firma seems about to be removed they are so much disturbed. To my own mind this is one of the most instructive and striking psychic manifestations of young animals, though I am not aware that any attention has been called to it before; and instead of referring to it under any of the usual divisions of sense, as the muscular sense, pressure sense, etc., I prefer to treat the subject under the above general heading (Sense of Support), for it seems to me that the feeling is a somewhat complex one.

"It is interesting to note that a water tortoise I have had for a number of years will at any time walk off a surface on which he is placed. But this is not a creature that always is on terra firma in the same sense as a dog, but it frequently has occasion to drop off logs, etc., into the water. But again, I find this sense of support well marked in birds which drop themselves into 'thin air'. Nevertheless, a consideration of ancestral experiences throws light on most cases, and perhaps on this one also."

Concerning white rats, SMALL ('99, p. 93) states that "as early as the second day (after birth) they show an uneasiness. when on the edge of a void—sometimes drawing back, sometimes manifesting their dominant trait of curiosity by leaning over and sniffing. At the age of four or five days the presence

of this sense (the sense of support) is unmistakable, and is not due to experience, as I have found by trying rats that have had no such experience." And WATSON ('03, p. 40) remarks that a rat that wanted to get down from the top of the food box "would usually stretch his head down two or three times, then pull himself back, as though he feared to attempt such a dangerous a feat."

The observations quoted indicate the possibility of interesting studies of the development of space perception in animals, and of such analyses of the sensory complex as shall exhibit the 'spatial worth' of each kind of sense data. Partly for the purpose of making an approach to the comparative study of space perception, partly for the solution of the following specific problems I have observed the behavior of several species of tortoises with respect to spacial conditions. The question which really led to the investigation was, What relation do the reactions of tortoises to space bear to their habits? Does the water species behave in essentially the same manner as does the landinhabiting form? The attempt to answer this question led to the study of the general behavior of different species, and of the importance of vision and the 'sense of support' in reactions to space.

Relation of Reactions to Space to Habits in Tortoises.

My method of experimentation was to place a tortoise in the middle of a board 30 cm. by 60 cm. which was elevated 30 cm., 90 cm. or 180 cm. above a net of black cloth into which the animal fell when it crawled or plunged over the edge of the board. The fall was thus rendered harmless to the animals, and they gave no evidence, by increased hesitancy in crawling off, that it was disagreeable to them. The observer carefully noted the behavior of the tortoise while it was on the board, and recorded the time that it remained there. It would seem that the time from the noticing of the edge of the board till the fall should be recorded rather than the total time spent on the board, but as it was found that some species notice the spatial conditions while they are still in the middle of the board, whereas

others give no evidence of perception of the height until they have reached the edge, it was necessary to make the record as described. Since in these experiments it was necessary that time as well as space should be considered, 60 minutes was fixed as the duration of the experiment, and in case the animal remained on the board longer than that period the test was recorded as a failure. Failures in this case have positive value, to be sure, but they do not give us the accurate measurement of the time of reaction which indefinite prolongation of the period of observation would furnish.

For detailed study three species were chosen: Chrysemys picta Schneider, as a representative of the water inhabiting forms; Nanemys guttata Schneider, to represent those species which spend part of their lives in water and part on land, and Terrapene carolina Linnaeus, as a strictly land inhabiting form.

Several individuals of each of the species were studied. In the tables the results for four individuals of each are pre sented. Each individual was given one trial a day at each of the three heights, 30, 90 and 180 cm. for ten days. In Table I we have a summary of the results, which are given in detail. for the various individuals in Table II. From an examination of the records the following facts appear: (1) The time spent on the board is shortest for the water species, longest for the land species. This indicates that the hesitation in the presence of a void increases as we pass from the strictly water forms to

TABLE I.

Reactions to Spatial Conditions of Tortoises of Different Habits. Summary of Results.

those which are land inhabiting; (2) Total inhibition of the reaction, i. e., failure to crawl over the edge of the board in the 60 minutes, appears at a much less height for the land species than for the water-land and water forms.

This quantitative expression of the amount of hesitation exhibited by different species of tortoises under the same spatial conditions clearly indicates a close relation between the demands of the natural environment of the species, so far as spa